KR101810279B1 - Real-time personalized recommendation of location-related entities - Google Patents

Abstract

In response to a query, a technique is described for ranking a location-based entity to show the user the ranking of location-based entities near the desired location. The technique can estimate popular location-based entities at specific locations and time zones, utilizing searches performed by a community of mobile device users at a given location and at a given time. The technique may also filter or re-rank these location-based entities based on the personal preferences of the person submitting the query and the group preferences of the group to which the user belongs. Also, personal preferences may vary depending on the situation (e.g., different restaurants preferred at lunch or in the evening).

Description

REAL-TIME PERSONALIZED RECOMMENDATION OF LOCATION-RELATED ENTITIES [0002]

Currently, when a user performs a search for a local company using a search engine, the ranking of the search results is static. That is, the search results depend only on location, and do not depend on the time the query was submitted, the person who submitted the query, and other information about the search. Such information of the situation may include, for example, weather, traffic conditions, popularity of the business, and the like. For example, a localized search result for a query "restaurant" at a given location does not change regardless of when the query was submitted. However, it is common for users, especially those submitting queries from mobile devices, to find other types of restaurants in the morning, at lunch or in the evening. In addition, different types of restaurants may be preferred at different times of the day, depending on the user's personal preferences.

Searches for local businesses and other areas of interest, particularly searches from mobile computing devices such as smartphones, reflect the user's interest in "right here, right now ". However, conventional techniques for ranking local information search results fail to capture such temporal and contextual dynamics.

summary ( Summary )

This summary is provided to introduce in simplified form selected concepts that will be described later in the description of the invention. This Summary is not intended to identify key features or key features of the claimed subject matter and is not to be used to limit the scope of the claims.

The location-based entity ranking technique described herein is a technique for ranking location-based entities such as, for example, local businesses, restaurants, entertainment venues, events, and the like. To this end, one embodiment of the present technique is to provide a mobile search log (record of search performed on a mobile computer device or the like) for ranking the rank of a location-based entity in real-time or near real- . Each time a user submits a query, the technique examines a location-based object among selected search results after other nearby users submit the same or similar query. In one embodiment, the technique only includes a portion of the mobile search log corresponding to a given time period. Additionally, in one embodiment of a technique for ranking location-based entities, there are two options for searching for location-based entities in response to a search query: real-time search and semi-real-time search.

In one embodiment of a real-time search of the technique, for example, to answer a given query submitted at a time T of a Monday, the technique uses a query submitted between the times (TX) and T on the very day Only. The variable X can be determined based on the amount of the query. The technique may increase the value of X (for example, by 1 hour, 2 hours, 3 hours, etc.) so that there are enough queries to perform the ranking.

In an embodiment of another quasi-real-time search, if the aforementioned real-time search option fails (e.g., the amount of available query is insufficient), similar time periods (e.g. Monday) ) Can be used to provide search results.

Regardless of how to identify the set of search logs, in one embodiment of the present technique, the more the location-based objects are selected from among the search results for the selected set of query logs, the higher the ranking score of the location-based objects . As the user selects (e.g., clicks on an input device) another location-based object at another time of the day, the position-based object's ranking and the search results that are visible are displayed in a variety of contexts such as weather, traffic conditions, It will change to reflect the limitation. In one embodiment, any ranking algorithm can be used to rank the location-based entities.

The technique of ranking a location-based entity utilizes a search performed by a community of computer device users, such as a mobile device user, at a given location and at a given time, to guess popular location-based objects in a particular location and time zone . The technique may then also filter or reorder these location-based entities based on the personal preferences of the user submitting the query. Personal preferences may also vary depending on the particular circumstances (eg, the preferences of the restaurant for lunch and dinner).

The particular features, aspects and advantages disclosed herein will be better understood from the following detailed description, appended claims and drawings.
FIG. 1 is an exemplary architecture that utilizes one exemplary embodiment of a ranking technique for location-based entities described herein.
Figure 2 illustrates a flow diagram of an exemplary process utilizing one embodiment of a technique for ranking a location-based entity.
3 illustrates a flow diagram of another exemplary process utilizing another embodiment of a technique for ranking a location-based entity.
4 illustrates an exemplary scenario using the ranking technique of position-based entities described herein.
Figure 5 illustrates an exemplary computer device that may be used to implement a ranking-based entity ranking technique.

In the following detailed description of a technique for ranking a location-based entity, reference may be made to the accompanying drawings, which are incorporated in and constitute a part of the specification, in which a ranking-based entity ranking technique described herein may be practiced Illustrate the invention and form part of it. It should also be understood that other embodiments may be utilized or structural changes may be made without departing from the scope of the invention.

1.0 Position-based object ranking technique ( Location - Related Entity  Ranking Technique )

In the following sections, an introduction and overview of a technique for ranking a location-based entity and an exemplary architecture and exemplary process for implementing the technique are provided. An exemplary scenario for applying this technique is also provided.

1.1 Introduction Introduction )

Contextual information is important when ranking search results of search results for location-based objects, such as queries for local businesses, restaurants, events, museums, parks or other areas of interest. However, it is difficult to obtain such information, annotate it, and classify it. Users themselves are a good source of such information. Various embodiments of the technique for ranking the location-based entities described herein utilize this point. By observing what each member of the community of people does after performing a search at a certain time of day and then choosing which one to select, the technique can be used to determine the popularity of that community under given circumstances (eg, temporal, personal, environmental) Although it is implicit, we can get a clear estimate of what it is.

1.2 Overview of this technique ( Overview of the Technique )

The technique of ranking the location-based entities described herein ranks the location-based entities according to their popularity. To this end, in one embodiment, the technique utilizes a search log (e.g., a record of searches done on a mobile device) of the mobile device to rank the location-based entities in real-time (or semi-real-time). (It should be noted that search logs by non-mobile computing devices may also be used if location information is available.) Additionally, search log data may be augmented with other data, such as phone log data or social networking data . For example, a phone call or post to a location-based object, like a search log, can be used to determine how often the location-based object is selected. In addition, in one embodiment, during ranking, a phone call or social networking post may be given a higher weight, since user selection of search logs provides implicit user preferences, . Each time a user submits a query, the technique examines a location-based entity in the search results that other users nearby select (e.g., click, call, or blog against) after submitting the same or similar query. The similarity of the query can be determined, for example, by comparing the similarity of the query terms. For example, if a user selects the same entity or homogeneous entity, the technique can compare the search term entered by the user with another search term input by other users. These search terms can be categorized and used as lexical markers when new search terms appear. So over time, this technique will relate the search terms "noodles" and "spaghetti" to search terms that produce quite similar results. For example, if the query "pasta" and "shrimp" all resulted in a 90% probability of choosing an Italian restaurant, then one embodiment of the technique would be similar in that two queries refer to the same type of business . In one embodiment, the invention may include only a portion of the mobile search log corresponding to a given time period.

In one embodiment of this technique for ranking location-based entities, there are two options for selecting a time zone for recommending a location-based entity to a user. The first option involves a real-time search for location-based entities of a type that the user may be interested in in response to the user's query. In one embodiment, the technique determines the type of location-based entity as follows. This technique examines all the location-based entities (for example, businesses) in the vicinity and calculates each score. The higher the score, the higher the position-based entity appears in the search results. This score is calculated based on the situation-limited log. For example, for a given query "Italian", the log will have a very small number of clicks on Japanese restaurants (if any). As a result, Japanese restaurants will receive low scores and will not appear at the top of the search results list. If this technique is to answer queries that have never been seen before, then this technique will show no results at all or show irrelevant results. The second option involves a quasi-real-time search for location-based entities. In this scenario, the technique searches for location-based entities for time periods that are similar (but not identical) to the time the query was entered.

Regardless of how to identify the set of search logs, the more a location-based object is selected by the user (e.g., clicked as an input device) from the search results for a selected set of query logs, The higher the ranking score is. If a user selects a different location-based object at different times of the day, the position of the location-based object and the search results that will be visible will vary, reflecting the cumbersome situational constraints of weather, traffic conditions, and the like. It should be noted that any ranking algorithm may be used to rank position-based entities. It should be noted that the scheme chooses a set of input data (e.g., search logs) to be input to the ranking algorithm (e.g., real-time option or quasi-real-time option) is important to the application of this technique.

In one embodiment, the technique finds personal preferences that vary with the time of a particular user submitting a query and filters and ranks the found location-based entities through the mobile search log using preferences that vary over time. Determining the personal preferences that vary over these times may be performed by monitoring the types of location-based entities the user selects after each query. In this way, the technique can learn over time the types of location-based entities that this user is interested in. In one embodiment, the technique may generate a user's preference based on different time of day or multiple days or different time periods within several seasons (e.g., summer versus winter). For example, if a user is known to like an Italian restaurant for lunch, then when the user searches for "restaurant" at lunch time, the technique may rank more in nearby Italian restaurants . If the user prefers a simple diner for the evening, the exact same query from that user during the evening will include a location-based entity, which is now a more streamlined restaurant.

In one embodiment, the technique may also utilize a user's personalized selection of search results to group users into communities or groups that are implicitly defined by their background, preferences, demographic characteristics . As a result, one embodiment of the present technique may modify the previously described search log analysis for a particular community of different rankings. For example, if a person who chooses an Italian restaurant in the evening usually chooses a pizza for lunch, and a given user sends a restaurant query around noon, the modified search log analysis will show that the nearby pizza restaurant is high I will mark it as having a ranking. Since the user's preferences change over time, the groups formed also vary over time. For example, a user can belong to a different group depending on the time the query was submitted.

In essence, in one embodiment, the technique utilizes searches performed by a community of mobile device users at a given location and time to guess popular location-based entities at particular locations and time zones. The technique then filters or reorders these location-based entities based on the personal preferences of the user who submitted the query. Personal preferences may be due to specific circumstances (eg, preferring other restaurants for lunch and dinner).

Since an overview of location-based ranking techniques has been provided, the following sections provide exemplary architectures and exemplary processes for implementing the techniques.

1.3 Example Architecture ( Exemplary Architecture )

Figure 1 provides a diagram of an exemplary architecture in which techniques for ranking the location-based entities can be implemented. 1, when the user 102 submits the query 104 using the mobile computer device 500 (discussed in more detail in FIG. 5), the query string 104 is returned to the user's location 106 (Or one or more servers that perform the same functions as those shown in block 110) along with user ID 108 (or optionally other relevant locations) and user ID 108. [ The query 104, the location 106 and the time 112 when the query 104 is submitted are recorded in the server 110. [

Based on location 106, time 112, and query 104 itself, portions of the set of search logs 114 (e.g., mobile search logs) are identified using real-time or quasi-real-time methods discussed below . In one embodiment of the present position-based entity ranking technique, there are two options for determining the portion of search log data to be used to recommend the location-based entity to the user. The first option, in response to a user's query, involves a real-time search for a type of location-based entity that the user may be interested in. The second option involves a quasi-real-time search for location-based entities. In this scenario, for example, the technique performs a search on a location-based entity over the course of a few days in the same time zone as the query was entered.

For example, in a real-time search option, for example, to answer a given query submitted at a time T of one Monday, the technique may, in one embodiment, calculate the time between the times TX and T on exactly the same day Only the submitted queries are used. The variable X can be determined according to the amount of the query. The technique may increase the value of X (e.g., by 1 hour, 2 hours, 3 hours, etc.) to ensure a sufficient number of queries to rank. In the semi-real-time search option, if the real-time search option described above fails (for example, the amount of available queries is insufficient), records within a given time zone (TX, T) Only the query logs are used.

A portion of the identified search logs 114 is used to rank the location-based entities within the module 116 that ranks the location-based entities of the server 110. The portion of search logs 114 used may be selectively filtered based on the user's group membership prior to ranking. This will be discussed in more detail in FIG. 3 and FIG. Group membership for a given user may change as the time period and location change.

Each time a user submits a query, the technique counts location-based entities in the selected search results after other nearby users submit the same or similar query. The more frequently a location-based entity appears in a selected portion of the search log, the higher the rank score. In practice, any ranking algorithm can be applied to this step.

In one embodiment of the present technique, based on the user ID, the personal preferences of the user submitting the query may be stored in the personalized database 118 and stored in a location-based entity ranking module 116, Can be further used to filter or reorder based objects. In addition, personal preferences can be adjusted according to time information (e.g., time, date, season, etc.). For example, a person's preference may be different when he submits a query on Monday morning.

More specifically, the user can provide personal preferences explicitly or implicitly. In the first case, in one embodiment, the user answers a question about his preferences (e.g., immediately after he installs the application for the technique on his mobile computing device). These personal preferences can change over time. The information provided by the user may be in the form of "I like expensive Italian or French restaurants on weekends", "I am usually interested in selling fast food on weekdays", "I prefer a place with outdoor seating" . Alternatively, the determination of user preferences may be implicit. For example, if a user chooses an expensive Italian or French restaurant with a 90% probability after submitting a "restaurant" query, the technique can assume that the user likes expensive Italian or French restaurants.

Many categories of location-based entities can be established. For example, these categories may include business attributes (e.g., mood, dress code, etc.), where users typically go to visit a given type of business, and the time when users search for or visit the business. Once the user preferences have been established, group preferences can be created by clustering user preferences together based on how similar the individual's preferences are.

Although real-time ranking of the location-based entity described herein is described as being performed on the server 110, such ranking may instead be performed on the mobile computer device 500 with a similar ranking module installed . In other embodiments, the functions described herein in connection with a server may be performed in a computing cloud.

The ranked ranked search results 120 returned in response to the query 104 are sent back to the user 102. After the user 102 selects a location-based entity from the returned search result 120, this data is written to the data logging module 124 of the server 110 and a new entry is created in the mobile search log 114 And the user's personal preferences are updated appropriately in the personalized database 118. The data logging module is responsible for recording user activity (e.g., in this case, a user's selection or click) in response to the query. In one embodiment, the data logger records all user selections (e.g., clicks) to ensure that this information is up-to-date. For example, a data logger records queries, user IDs, device types, hours of the day, days of the week, and so on. This information is used to perform ranking of location-based entities.

The real-time ranking module 116 periodically reorders the local location-based objects for each community, time zone, and other context parameter values to reflect user preferences in real time. For example, in one embodiment, as described above, each time the user clicks (selects) a predetermined search result, the search log is updated. Since this is after real-time ranking, the technique attempts to capture changes in user selection (e.g., click) manner over time. To do so, the technique periodically calculates the rank of a local location-based entity (e.g., a business) for every combination of variables of the situation under consideration (e.g., time of day, day of week, etc.). By doing this periodically, this technique makes it possible to rank in real time (because the rankings are already calculated) and allow the server to respond quickly to queries.

Having discussed an exemplary architecture for implementing this technique, the following sections describe some exemplary processes for implementing this technique.

1.4 Exemplary process using position-based object ranking technique Processes for Employing the Location - Related Entity Ranking Technique )

A flow diagram of an exemplary embodiment 200 of performing a ranking-based entity ranking technique is shown in FIG. As can be seen in Figure 2, at block 202, a query for the type, location, and location of a location-based entity (e.g., business, location, area of interest) and a given time do. Then, as shown in block 204, the search log of the time zone associated with the input time is retrieved and the location of the type of input location-based entity selected by the user from the search log around the input location in response to the same or similar query - Finds a base object. The location-based entity selected by the user as shown in block 206 is then ranked according to the number of times each location-based entity is selected by the user from the search log. The ranked list of location-based entities is then output as shown in block 208.

A flow diagram of another exemplary embodiment 300 for performing a ranking-based entity ranking technique is shown in FIG. As shown in FIG. 3, at block 302, a query for a location-based entity having a given type is entered into the search engine at a given time and location. At block 304, a search log from some users associated with the input query is also input. These search logs may also be filtered (or optionally filtered to include a time zone similar to the input time) to include only a time period around a given input time. The search log associated with the search query may be filtered based on group membership preferences (block 306) as discussed above in FIG. 2 above.

Searches these filtered search logs in the time zone associated with the input time to find a location-based object of the same type as the location-based object input selected by the user from the filtered search log near the input location in response to the same or similar query (Block 308). The location-based entities selected by the user are ranked by the number of times the respective location-based entities are selected by the user from the filtered search logs, as shown in block 310.

The ranked list of location-based entities is then re-ranked based on the user preference of the user who entered the query, as shown in block 312, and the output of the location- do. Then, to update the search log and preferences, the user who entered the query, as shown in block 316, selects the location-based entity from the ranked list of location-based entities (block 318) The search log and the user preferences are updated to show that the user has selected the selected location-based entity in response to the query.

1.5 Exemplary Scenarios Using Location-Based Object Ranking Techniques Scenario for Using the Location - Related Entity Ranking Technique )

In FIG. 4, an exemplary scenario 400 is presented that illustrates how to rank a location-based entity in one embodiment of a ranking-based entity ranking technique. To simplify the scenario, suppose that the lowest possible location resolution is the zip code unit, and the user is interested in ranking the location-based objects within the one hour time zone. In practice, any available position unit and resolution may be used and any time zone may be used.

As shown in Figure 4, when a user (denoted by user ID 402) is at time 408 of "local time 9 o'clock" at location 406 where query "restaurant" 404 is identified with zip code "98004" . When this information 404, 406, 408 reaches the server 410, the server logs in to the postal code "98004" that was recorded in the past time (8-9 pm) Will utilize real-time search indexes that reflect all of the entries in the corresponding mobile search logs 412. If the user has a known profile and belongs, for example, to the "person who likes restaurants" category, the server uses the search index built for that particular user group 416 using the filtering module 414 . Given an identified portion of the mobile search entry 418, the ranking procedure is applied to the ranking module 420 to rank the location-based entities selected by other similar users. In this example, "Maggiano's" gets 3 points because it appears in 3 entries, and "Subway" gets 1 point because it appears in only one entry. The total number of search log entries used to perform ranking of location-based entities can be used to evaluate the reliability of ranking results (the more number of entries used, the more reliable the ranking results). The ranked position-based entity is further filtered in the filtering / re-ranking module 424 using a personal preference 422 that varies with the time of the user who submitted the query. In this scenario, "It appears to be interested only in Italian and Mexican restaurants (block 426). As a result, the "Subway " is excluded from search results 428 as an unrelated business for a particular user (or alternatively may be ranked low for that user). It should be noted that there may be multiple personal preference tables for different time periods or days, even for the same user.

2.0 Computing Environment ( The Computing Environment )

The technique of ranking location-based entities is designed to operate in a given computing environment. The following description is intended to provide a brief, general description of a suitable computing environment in which a ranking-based entity ranking technique may be implemented. The techniques are operable in a variety of general purpose or special purpose computer system environments or configurations. Examples of well known and appropriate computing systems, environments and / or configurations include, but are not limited to, personal computers, server computers, hand-held or laptop devices (e.g., media players, notebook computers, cellular phones, personal data assistants, A distributed computing environment including any one or more of the above systems or devices, etc.), multiprocessor systems, microprocessor-based systems, set top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, , But is not limited thereto.

5 illustrates an example of a suitable computing system environment. The computing system environment is merely an example of a suitable computing environment and is not intended to suggest limitations on the scope of use or functionality of the techniques. Also, the computing environment should not be construed as having any dependency or requirement relating to the components or combinations thereof shown in this exemplary operating environment. Referring to FIG. 5, an exemplary system for implementing a technique for ranking a location-based entity includes a computing device, such as computer device 500. In the most basic configuration, the computing device 500 typically includes at least one processing unit 502 and a memory 504. Depending on the exact configuration and type of computer device, memory 504 may be volatile (such as RAM), non-volatile (such as ROM or flash memory, etc.), or a combination thereof. This most basic configuration is shown by dashed line 506 in Fig. Additionally, the device 500 may also have additional features / functions. For example, the device 500 may include, but is not limited to, a magnetic or optical disk or additional storage devices such as tape (removable and / or stationary). This additional storage device is illustrated in Figure 5 by a removable storage device 508 and a fixed storage device 510. Computer storage media includes volatile and nonvolatile, removable and non-removable media in which methods or techniques for storing information such as computer readable instructions, data structures, program modules or other data are implemented. The memory 504, the removable storage 508, and the fixed storage 510 are all examples of computer storage media. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, Or any other medium which can be used to store the desired information and which is accessible by the apparatus 500, but is not limited thereto. The computer-readable medium includes both temporal propagation signals and computer (readable) storage media. Any computer storage media may be a part of the apparatus 500.

The device 500 may also include communication interface (s) 512 that allow the device to communicate with other devices or networks. The communication connection (s) 512 is an example of a communication medium. Communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media. The term "modulated data signal" means a signal that has one or more characteristic sets or that is modified in such a way as to encode information in the signal, which changes the configuration or state of the device receiving the signal. By way of example, and not limitation, communication media includes wired media such as a wired network or direct-wired connection and wireless media such as acoustic, RF, infrared, or other wireless media. The term computer readable medium as used herein includes both storage media and communication media.

The device 500 may include various input device (s) 514, such as a display, keyboard, mouse, pen, camera, touch input device, Output device (s) 516, such as a display, speaker, printer, etc., may also be included. All of these devices are well known in the art and need not be discussed here long.

Techniques for ranking location-based entities can be described in the general context of computer-executable instructions, such as, for example, program modules executed by a computer device. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular types of abstract data. The technique of ranking location-based objects may be implemented in a distributed computing environment where tasks are performed by remote processing devices connected through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices. In addition, the above-described instructions may be implemented in whole or in part by a hardware logic circuit that may or may not include a processor.

It should be noted that one or all of the above-described embodiments described herein may be used in any desired combination to form additional hybrid embodiments. Although the invention has been described in language specific to structural features and / or methodical acts, the invention as defined in the appended claims is not necessarily limited to the features or acts described above. The above-described features and operations are disclosed as exemplary forms of implementing the claims.

Claims (20)

A computer-implemented method for ranking location-related entities in a particular geographic area in response to a search query,
Receiving a query for a geographic location-based entity having a given type, a location within a particular geographic area, and a current time;
In response to a query that is the same as or similar to the received query, a mobile search log of the mobile computing device for location-based entities in the particular geographic area with the received given type selected by users near the received location Retrieving, using the received query, a real time window of time including the current time;
In response to a query that is the same as or similar to the received query in the real-time band that includes the current time, each geo-location-based entity is allowed to send to the users from the mobile search log according to the number of times selected by the users from the mobile search log Ranking the geographic location-based entities within the specific geographic area selected by the geographic location-based entities;
Outputting to the user a ranked list of the geographical location-based entities within the particular geographical area most frequently selected by users in the real-time band including the current time
≪ / RTI >
The method according to claim 1,
Based on the time-dependent user preference of the user who entered the query, before outputting a ranked list of the location-based entities, Further comprising the step of re-ranking
A computer implemented method.
3. The method of claim 2,
The user preference may vary depending on the location of the user
A computer implemented method.
The method according to claim 1,
Determining time-dependent group memberships by grouping users by time, location, query, and query result
A computer implemented method.
5. The method of claim 4,
Further comprising filtering the mobile search log based on time-dependent group membership preferences prior to retrieving the mobile search log
A computer implemented method.
6. The method of claim 5,
The groups to which the user belongs can be varied according to different time periods and locations
A computer implemented method.
The method according to claim 1,
The real time zone is selected based on the time at which the query was entered
A computer implemented method.
8. The method of claim 7,
If the mobile search log data does not exist to generate the ranked list in real time, the real time zone associated with the current time is selected to be within the same time zone as the input time of the previous day
A computer implemented method.
The method according to claim 1,
The step of ranking the geographical location-based entities comprises:
A computer implemented method.
A computer-implemented method for ranking location-based entities,
Receiving a query for a location-based entity type at a current time and a geographic location within a particular geographic area,
Receiving a mobile search log of a mobile computing device associated with the received query and also based on a query volume and localized to a time zone adjacent to the current time;
A location within the particular geographic area having the received location-based entity type selected from the mobile search log by users near the input location in response to a query the same or similar to the received query at a time zone adjacent to the current time, Retrieving the mobile search log for base entities using the received query at a time zone adjacent to the current time;
Wherein each location-based entity type in the time zone adjacent to the current time ranks the location-based entities in the particular geographic area selected by users according to the number of times selected by the users from the received mobile search log Step,
Outputting a ranked list of said location-based entities in said particular geographic area most frequently selected by users in the time zone adjacent to said current time
≪ / RTI >
11. The method of claim 10,
Further comprising ranking the ranked list of the location-based entities based on a time-dependent user preference of the user who entered the query
A computer implemented method.
11. The method of claim 10,
Further comprising filtering the mobile search log based on time and location-dependent group preferences of the group to which the user who entered the query belongs before searching the mobile search log
A computer implemented method.
11. The method of claim 10,
Selecting a location-based entity from the ranked list by a user entering the query;
Updating the mobile search log to show that the user has selected the selected location-based entity in response to the query
Lt; RTI ID = 0.0 > 1, < / RTI >
11. The method of claim 10,
Selecting a location-based entity from the ranked list of ranked location-based entities by the user who entered the query;
Updating the user's time-dependent user affinity to reflect the location-based entity selected by the user
Lt; RTI ID = 0.0 > 1, < / RTI > 15. The method of claim 14,
The updated time-dependent user preference
A location that the user normally visits,
The type of location the user visits, and
A distance at which the user is allowed to move to visit a predetermined position
≪ / RTI >
A system for determining a popular location-based entity around a given location at a given time,
A computing device,
A computer program comprising a program module executable by the computing device,
The computing device may be programmed by a program module of the computer program,
Receiving a query for retrieving a mobile search log of a mobile computing device for a type of geographical location-based entity at a given location within a particular geographical area at a current time;
Filtering the mobile search log based on a group membership of a user who has entered the query;
For the type of geographic location-based entity at the given location selected from the filtered mobile search log by users near the input location in response to a query that is the same as or similar to the received query at a time zone adjacent to the current time Retrieving the filtered mobile search log;
Based on the number of times selected from the filtered mobile search log in a time zone adjacent to the current time in response to a query that is the same as or similar to the input query, And
Receiving, by the user, a ranked list of geo-location-based entities in the geo-area most frequently selected by users in a time zone including the current time,
A system for determining a popular location-based entity around a given location at a given time. 17. The method of claim 16,
Further comprising a module for re-ranking the ranked list according to a preference of a user who entered the query that varies with time during the day
A system for determining a popular location-based entity around a given location at a given time. 17. The method of claim 16,
Further comprising a module for re-ranking the ranked list based on time-of-day time-varying group preferences
A system for determining a popular location-based entity around a given location at a given time. 17. The method of claim 16,
Further comprising a module for using an additional data source to increase said mobile search log data comprising telephone log data
A system for determining a popular location-based entity around a given location at a given time. 17. The method of claim 16,
Further comprising a module that uses an additional data source to increase the mobile search log data comprising social network data
A system for determining a popular location-based entity around a given location at a given time.

Images